Sympathetic Enhancement of Memory T-Cell Homing and Hypertension Sensitization

Abstract

Rationale: Effector memory T lymphocytes (T_EM cells) exacerbate hypertension in response to repeated hypertensive stimuli. These cells reside in the bone marrow for prolonged periods and can be reactivated on reexposure to the hypertensive stimulus.

Objective: Because hypertension is associated with increased sympathetic outflow to the bone marrow, we hypothesized that sympathetic nerves regulate accumulation and reactivation of bone marrow-residing hypertension-specific T_EM cells.

Methods and results: Using unilateral superior cervical ganglionectomy in wild-type C57BL/6 mice, we showed that sympathetic nerves create a bone marrow environment that supports residence of hypertension-specific CD8⁺ T cells. These cells, defined by their proliferative response on coculture with dendritic cells from Ang (angiotensin) II-infused mice, were reduced in denervated compared with innervated bone of Ang II-infused mice. Adoptively transferred CD8⁺ T cells from Ang II-infused mice preferentially homed to innervated compared with denervated bone. In contrast, ovalbumin responsive T cells from OT-I mice did not exhibit this preferential homing. Increasing superior cervical ganglion activity by activating Gq-coupled designer receptor exclusively activated by designer drug augmented CD8⁺ T_EM bone marrow accumulation. Adoptive transfer studies using mice lacking β2AR (β2 adrenergic receptors) indicate that β2AR in the bone marrow niche, rather than T-cell β2AR is critical for T_EM cell homing. Inhibition of global sympathetic outflow using Gi-coupled DREADD (designer receptor exclusively activated by designer drug) injected into the rostral ventrolateral medulla or treatment with a β2AR antagonist reduced hypertension-specific CD8⁺ T_EM cells in the bone marrow and reduced the hypertensive response to a subsequent response to low dose Ang II.

Conclusions: Sympathetic nerves contribute to the homing and survival of hypertension-specific T_EM cells in the bone marrow after they are formed in hypertension. Inhibition of sympathetic nerve activity and β2AR blockade reduces these cells and prevents the blood pressure elevation and renal inflammation on reexposure to hypertension stimuli.

Keywords: angiotensin II; dendritic cells; ganglionectomy; hypertension; inflammation.

Figure 1:. Sympathetic innervation of the bone marrow and the effects of superior cervical ganglionectomy (SCGx).

(A) Western blot analysis of tyrosine hydroxylase (TH) expression in the forelimb bone marrow in mice with two weeks of sham or angiotensin II infusion. Data are expressed as mean ± SEM (Sham: 1.000±0.086 vs. Ang II: 2.065±0.224), n=14 in each group, p=0.0004 for the effect of angiotensin II was calculated by unpaired t test with Welch’s correction. (B) Example of ptosis resulting from unilateral SCGx. (C) Western blot analysis of TH expression in the innervated (In.) and denervated (De.) bone marrow. Protein samples were extracted from forelimb bone marrow of humerus, ulna and radius, and β-tubulin was probed as a loading control. Each set of connected symbols represent paired bone marrow samples from the same animal, n=8, p=0.0013 for effect of denervation analyzed by paired t test. (D) Expression of TH and its co-localization with the vascular marker CD31 were detected in the calvaria from mice that had received unilateral SCGx by confocal fluorescence microscopy. White bars indicate 100 micrometers. **p<0.01, ***P<0.001

Figure 2:. Effects of sympathetic nerves on the accumulation of hypertension-specific T cells in the bone marrow.

(A) Dendritic cells isolated from sham or angiotensin II infused mice were cultured with bone marrow cells obtained from either denervated or innervated forelimbs of other hypertensive mice. Bone marrow cells were pre-labeled with CFSE. After 7 days in culture, CD3⁺, CD4⁺, CD8⁺ T cells were quantified by flow cytometry (B to D). Differences in the proliferation of CD8⁺ T cells from control and denervated bone marrow were determined by CFSE dilution and flow cytometry (E). Each set of connected symbols represent paired bone marrow samples from the same animal, n=5 to 6 in each group, p<0.0001 for the effect of denervation calculated by paired t test is shown.

Figure 3:. Effects of sympathetic nerves on the accumulation of CD8⁺ effector memory T cells in the bone marrow after hypertension.

(A) Splenic pan-T cells were isolated from hypertensive wild type CD45.2 donor and adoptively transferred to CD45.1 recipient that either had unilateral SCGx, or AAV expressing either a control or Gq-DREADD injected into the SCG. In the case of the Gq-DREADD experiments, Clozapine-N-oxide (CNO) was administered in the drinking water for a week after adoptive transfer. One week later, recipient forelimb bone marrow was analyzed by flow cytometry. (B) A representative sample showing the gating strategy of central memory T cells (T_CM) and effector memory T cells (T_EM) in both CD4⁺ and CD8⁺ population from donor mice. CD8⁺ T_EM cells are emphasized in red. After adoptive transfer, CD45.2⁺/CD8⁺ T_EM cells were detected in the recipient mice with unilateral SCGx or those that had unilateral Gq-DREADD activation in SCG. The cells were quantified respectively in panel C (Innervated vs. Denervated, p=0.0073) and panel D (control vs. Gq-DREADD, p=0.0009), n=7 in each experiment. Expression of mCherry tagged Gq-DREADD was detected by confocal fluorescence microscopy in the SCG-innervated bone marrow and was co-localized with sympathetic nerve marker tyrosine hydroxylase (TH) (panel E, White bars indicate 50 micrometers). Levels of norepinephrine (NE) and epinephrine (Epi) in the bone marrow were measured by HPLC (panel F), n=9 in both groups. Each set of connected symbols represent paired bone marrow samples from the same animal, p=0.0040 in norepinephrine and p=0.0255 in epinephrine for the effect of Gq-DREADD as calculated by paired t test. * P<0.05, **P<0.01.

Figure 4:. Role of β2 adrenergic signaling in the bone marrow homing of CD8⁺ effector memory T cells after hypertension.

(A) Pan T cells were isolated for the spleen of wild type (CD45.2) mice that had two weeks of either sham or angiotensin II infusion. Bone marrow cells were isolated from both innervated and denervated forelimbs of B6 Cd45.1 mice that had unilateral SCGx. (B) Accumulation of transmigrated T cells isolated from innervated (In.) and denervated (De.) bone marrow (BM). Data are expressed as mean ± SEM (Sham In. BM: 3.00±0.53 vs. Ang In. BM: 17.46±2.47, p=0.0003; Sham De. BM: 5.27±1.31 vs. Ang De BM: 20.00±2.96, p=0.0002), P values for the differences between each treatments was calculated by two-way ANOVA with repeated measurements followed by a Bonferroni post hoc test, n=5 in each group. (C) Accumulation of T cells from mice after angiotensin II infusion with bone marrow cells from naïve B6 Cd45.1 mice in the presence of no treatment, or 1 μmol/L norepinephrine (NE), norepinephrine (NE) and the β2 adrenergic receptor antagonist ICI118,551 (10 nmol/L) added 30 minutes before NE (ICI + NE), or the selective β2 adrenergic receptor agonist salbutamol (Sal, 1 μmol/L). P values for the differences between each two groups were calculated by one-way ANOVA with repeated measurement followed by a Tukey’s post hoc test, n=6 in each group. (D) Adoptive transfer of splenic pan-T cells was performed between Adrb2^−/− mice and B6 Cd45.1 mice. The donor received two weeks of angiotensin II infusion and the recipients received unilateral SCGx prior to the adoptive transfer. (E) CD8⁺ T_EM cells in the innervated and denervated forelimb bone marrow were quantified by flow cytometry. For experiements shown in panel C and E, each set of connected symbols represent paired bone marrow samples from the same animal, the p values for the effect of Adrb2 gene deficiency in T cells vs. bone marrow was calculated by two-way ANOVA followed by a Bonferroni post hoc test, n=4 to 8 in each group. ** P<0.01, *** P<0.001.

Figure 5:. Effects of sympathetic nerves on bone marrow chemokine expression.

Bone marrow samples were collected from both the innervated (In) and denervated (De) bones of mice one week after unilateral SCGx, and mRNA expression of C-C chemokine ligands (CCL)-19, CCL-21, vascular cell adhesion molecule 1 (VCAM-1) and intercellular adhesion molecule 1 (ICAM-1) were measured by real-time PCR. Each set of connected symbols represent paired bone marrow samples from the same individual animal. The p values for the effect of denervation were calculated by paired t test, n=5 in each group. * p=0.0128 in CCL19, ** p=0.0032 in CCL21a.

Figure 6:

Effects of sympathetic nerves on the accumulation of OT-I memory T cells in the bone marrow after immunization with the OVA257–264 (SIINFEKL) peptide. Splenic pan-T cells were isolated from immunized OT-I mice and transferred to B6 Cd45.1 recipient that had unilateral SCGx a week earlier. One week later, the recipient bone marrow in was analyzed by flow cytometry. (A) Representative sample showing the gating strategy. Total TCR Vβ5.1/5.2⁺ OT-I T cells, Vβ5.1/5.2⁺/CD8⁺ T cells as well as subsets of central and effector memory T cells in CD8⁺ T cell population were quantified by flow cytometry. Each set of connected symbols represent paired bone marrow samples from the same individual animal. No statistically significant difference was detected by paired t test, n=8 in each group.

Figure 7:

Effects of systemic sympatho-inhibition during memory T cell homing on future hypertension development in response to low dose angiotensin II infusion. (A) Experimental paradigm employed. (B) Bilateral rostral ventrolateral medulla (RVLM) microinjection targets shown in coronal section of brainstem (marked in red circles). White bars indicate 100 micrometers. (C) Three-day measurements of systolic blood pressure were obtained by radiotelemetry at baseline and during the first and second week of angiotensin II infusion. After two weeks of angiotensin II infusion, kidneys from mice was harvested and infiltrating inflammatory cells were quantified by flow cytometry. Mean data for total leukocytes (CD45⁺), total T cells (CD3⁺), CD8⁺ and CD4⁺ T cells are shown in (D) to (G). Central and effector memory T cell subsets in CD8⁺ and CD4⁺ cells were further quantified as shown in (H) to (K). Blood pressure data were analyzed with 2-way ANOVA with repeated measurements, p=0.0124 between the two groups during angiotensin II infusion, n=5 in each group. Flow cytometry data were analyzed by unpaired t tests and p values between two groups were calculated (CD45⁺: p=0.0102, CD3⁺: p=0.0248, CD8⁺: p=0.0051, CD4⁺: p=0.0005, CD8⁺ T_EM: p=0.0015, CD4⁺ T_EM: p=0.0014, CD8⁺ T_CM: p=0.4621, CD4⁺ T_CM: p=0.3216), n=7 and 8 in each group, *P<0.05, **P<0.01, ***P<0.001 in the figure.

Figure 8:

Effects of β2 blockade on bone marrow memory T cells and future hypertension development. (A) After 2 weeks of angiotensin II infusion at pressor dose (490 ng/kg/min), mini-osmotic pumps for angiotensin II infusion were removed, and the mice received another pump for either vehicle infusion or selective β2 adrenergic receptor antagonist ICI118,551 infusion (200 ng/kg/min) for two weeks. (B) In a subset of mice, bone marrow was harvested for T cell proliferation assay. After 7 days of co-culture with DCs, CD3⁺, CD4⁺, CD8⁺ T cells and two T_EM cell subsets in the bone marrow were quantified by flow cytometry as shown respectively from panels (C) to (G). CD3⁺: Veh: 20.5±0.8% vs. ICI:14.8±4.1%, p=0.0024; CD8⁺: Veh: 19.0±0.8% vs. ICI:14.0±1.4%, p=0.0031; CD4⁺: Veh: 1.08±0.06% vs. ICI:0.70±0.08%, p=0.0012; CD8⁺T_EM: Veh: 15.6±0.6% vs. ICI:10.1±0.9%, p<0.0001; CD8⁺T_EM: Veh: 0.91±0.17% vs. ICI:0.58±0.08%, p=0.0023. In another subset of mice, the ICI118,551 minipump was removed and a 3^rd minipump inserted to administer a subpressor dose of angiotensin II (140 ng/kg/min). BP was measured using radiotelemetry (panel H). Data are expressed as mean ± SEM. Flow cytometry data were analyzed by unpaired t tests, n=8 in each group. Blood pressure was analyzed using 2-way ANOVA with repeated measurements, n=5 in each group, p=0.030 between the two groups during angiotensin II infusion, n=5 in each group. *P<0.05, **P<0.01, ***P<0.001.